Performance of the MasSpec Pen for Rapid Diagnosis of Ovarian Cancer

Abstract

BACKGROUND Accurate tissue diagnosis during ovarian cancer surgery is critical to maximize cancer excision and define treatment options. Yet, current methods for intraoperative tissue evaluation can be time intensive and subjective. We have developed a handheld and biocompatible device coupled to a mass spectrometer, the MasSpec Pen, which uses a discrete water droplet for molecular extraction and rapid tissue diagnosis. Here we evaluated the performance of this technology for ovarian cancer diagnosis across different sample sets, tissue types, and mass spectrometry systems. METHODS MasSpec Pen analyses were performed on 192 ovarian, fallopian tube, and peritoneum tissue samples. Samples were evaluated by expert pathologists to confirm diagnosis. Performance using an Orbitrap and a linear ion trap mass spectrometer was tested. Statistical models were generated using machine learning and evaluated using validation and test sets. RESULTS High performance for high-grade serous carcinoma (n = 131; clinical sensitivity, 96.7%; specificity, 95.7%) and overall cancer (n = 138; clinical sensitivity, 94.0%; specificity, 94.4%) diagnoses was achieved using Orbitrap data. Variations in the mass spectra from normal tissue, low-grade, and high-grade serous ovarian cancers were observed. Discrimination between cancer and fallopian tube or peritoneum tissues was also achieved with accuracies of 92.6% and 87.9%, respectively, and 100% clinical specificity for both. Using ion trap data, excellent results for high-grade serous cancer vs normal ovarian differentiation (n = 40; clinical sensitivity, 100%; specificity, 100%) were obtained. CONCLUSIONS The MasSpec Pen, together with machine learning, provides robust molecular models for ovarian serous cancer prediction and thus has potential for clinical use for rapid and accurate ovarian cancer diagnosis.

The intelligent knife (iKnife) and its intraoperative diagnostic advantage for the treatment of cervical disease

“…In our case, there was more than an order of magnitude difference in the number of dimensions (d = 4,000) and the number of samples (n = 87). Although this is not ideal for most of the machine learning algorithms, LASSO has been used successfully in similar cases (i.e., ovarian cancers) (45). LASSO algorithm focuses on the features differentiating between the given classes, and thus it can be more accurate compared to PCA in cases when the model contains more classes and the deviation between the classes is not that significant.…”

Identification of diagnostic metabolic signatures in clear cell renal cell carcinoma using mass spectrometry imaging

“…31 The high accuracy of predicting cancer based on metabolites suggests that DESI-MS and mass spectrometry-based tools like a mass spectrometry pen that has been deployed intraoperatively in ovarian cancer with high accuracy for diagnosis. 32 DESI-MS can provide a rapid readout of diagnostic mass spectra that make it compatible with the goal of minimizing ischemia time and identifying PSM in the surgical bed. Such an enabling technology could facilitate safe expansion of indications for PN to more complex cases based on tumor size and location.…”

A lipidomic workflow capable of resolving sn- and CC location isomers of phosphatidylcholines

“…19 In this study, we were interested to test if changes of PC snisomers would differentiate lipidomes of diseased and healthy states, which may be potentially used as markers in biomedical and clinical studies. 46 Using the HILIC-LC-MS workow described earlier, we thus evaluated this idea by analyzing PCs in cancerous human breast tissue samples (N ¼ 3), while paracarcinoma tissue samples (N ¼ 3) were used as controls (three technical repeats were acquired for each sample). Four levels of structural and quantitative analysis were performed for PCs, including subclass, fatty acyl composition, sn-composition, and C]C location.…”